Text clustering method based on weak supervised deep learning

Abstract

The invention discloses a text clustering method based on weak supervised deep learning. The method comprises the following steps: (1) by means of an image data set with text click information, imagevisual information and image category labels are utilized, and adopting image amplification and clustering to construct an image category click characteristic matrix of each text; And (2) obtaining asmooth image click feature map on the initial class click matrix by using a sorting and propagation method. Performing text clustering on the feature map to obtain an initial text category, and initializing text weight by utilizing click priori; (3) under the condition of minimizing an intra-class mean square error, building a deep text clustering model to learn deep text characteristics; (4) performing joint optimization on the depth model and the text weight by using a weak supervised learning method, and iteratively updating the depth model and the text weight; (5) deep text features are extracted through the deep text model, and K-based text feature extraction is achieved. And clustering the means method. The method has very high universality, and the semantic gap in image recognitionis effectively solved.

Description

technical field [0001] The invention relates to the fields of fine-grained image classification and text clustering, in particular to a text clustering method based on deep learning with weak supervision. Background technique [0002] Text is a word, phrase or sentence that contains semantic information. In life and work, people often use text to express their thoughts and feelings. Exploding text information makes text clustering tasks especially important. The present invention proposes an improved text clustering method and applies this method in fine-grained image classification to learn compact text representations for images. This method solves the "semantic gap" problem of images to a certain extent and improves the accuracy of image recognition. [0003] Recently, researchers proposed to model the relationship between text and images using the click data class. Click data is images and corresponding text feedback information collected by search engines (such as G...

AI Technical Summary

Problems solved by technology

[0004] The present invention has carried out sufficient experiments on Microsoft's click data Clickture. The data set contains more than 480,000 texts. In these texts, wrong input, similar words, and irrelevant words account for a very large proportion. If manual cleaning and screening are considered, it is obviously costly. huge and unreasonable

Images